WO2010002630A2 - Computing higher resolution images from multiple lower resolution images - Google Patents
Computing higher resolution images from multiple lower resolution images Download PDFInfo
- Publication number
- WO2010002630A2 WO2010002630A2 PCT/US2009/048241 US2009048241W WO2010002630A2 WO 2010002630 A2 WO2010002630 A2 WO 2010002630A2 US 2009048241 W US2009048241 W US 2009048241W WO 2010002630 A2 WO2010002630 A2 WO 2010002630A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- resolution image
- tile
- processor
- pixels
- lower resolution
- Prior art date
Links
- 230000003287 optical effect Effects 0.000 claims abstract description 11
- 239000013598 vector Substances 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 description 37
- 238000005457 optimization Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical group O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000002939 conjugate gradient method Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 229910003460 diamond Inorganic materials 0.000 description 2
- 239000010432 diamond Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008707 rearrangement Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000008521 reorganization Effects 0.000 description 1
- 238000012857 repacking Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
Definitions
- This relates generally to generating so-called super- resolution images which are higher resolution images based on multiple lower resolution images.
- Images with higher resolution are more desirable. Images with higher resolution have greater pixel density and, hence, show greater detail than lower resolution images of the same scene. Higher resolution images have many applications, including medical imaging, satellite imaging, computer vision, video surveillance, face recognition, car plate number extraction and recognition, and converting digital versatile disk video to high density television, to mention a few examples.
- the lower resolution images may be different views of the same scene. They may be obtained from the same camera, for example, while introducing small, so-called sub-pixel shifts in the camera location from frame to frame, or capturing a small amount of motion in the scene.
- the low resolution images may be captured using different cameras aimed at the same scene.
- a resultant high resolution image is then reconstructed by aligning and properly combining the low resolution images so that additional image information is obtained.
- the process may also include image restoration, where de-blurring and de-noising operations are performed as well.
- the reconstruction of the resultant high resolution image is a difficult problem because it belongs to the class of inverse, ill-posed mathematical problems.
- the needed signal processing may be interpreted as being the reverse of the so-called observation model, which is a mathematically deterministic way to describe the formation of low resolution images of a scene based upon known camera parameters. Since the scene is approximated by an acceptable quality high resolution image of it, the observation model is usually defined as relating a high resolution discrete image of the scene to its corresponding low resolution images. This relationship may be given as a concatenation of geometric transform, a blur operator, and down-sampling operator, plus an additive noise term.
- Figure 1 is a flow diagram for a re-packing algorithm in accordance with one embodiment of the present invention
- Figure 2 is a flow chart for the gradient computation in accordance with one embodiment of the present invention
- Figure 3 is a flow chart for non-linear iterative conjugate gradient optimization in accordance with one embodiment.
- Figure 4 is a system depiction for one embodiment.
- the letter n represents additive noise having random characteristics, which may represent, for example, the variation or error between lower resolution images that have been captured by the same camera without any changes in the scene and without any changes to camera or lighting settings.
- the super-resolution image reconstruction estimates the higher resolution image f that corresponds to a given set of lower resolution images Y.
- a Bayesian estimation process may be used to estimate f, to get the resultant higher resolution image mentioned above.
- an "a posteriori" probability function typically a probability density function, is mathematically defined as P(f
- the optimization problem which is finding a suitable high resolution image f, that has the highest probability, given a set of low resolution images or that maximizes the a posteriori probability function, may be rewritten as: P(f
- Y) p(Y
- the prior probability density function indicates what higher resolution images are more likely to occur, based on, for example, statistical characterization of an ensemble of different higher resolution images.
- the prior probability density function may be a joint probability, defined over all the pixels in a higher resolution image and may be based on statistical data from a large number of images.
- the a posteriori probability function is a probability density function that defines the probabilities of observing lower resolution images that correspond to a particular higher resolution image.
- the likelihood probability density function may be determined based on the observation model described above where the noise term is typically assumed to have a Gaussian probability distribution.
- the estimation process becomes one of iteratively determining trial higher resolution images and stopping when there is a convergence, which may signify that a maximum of the a posteriori probability function has been reached.
- a non-linear iterative conjugate gradient optimization algorithm computes the gradient of the a posteriori probability function at each search iteration.
- the calculation of the gradient is the most computationally intensive part of the super-resolution algorithm. In typical configurations it takes 80 to 90 percent of the maximization procedure running time .
- the conjugate gradient optimization algorithm models noise by a probabilistic, non-Gaussian, robust function.
- Figure 3 shows the use of the conjugate gradient method which is an iterative method that provides an acceptable balance between complexity and speed of convergence.
- the criteria for convergence is delta E less than T (diamond 48), which tests whether the error or difference in the a posteriori probability function between two successive trial higher resolution images, is less than a predefined threshold T.
- An alternative is to define delta E as the difference between consecutive trial higher resolution images .
- the conjugate gradient method computes the gradient of the error function which has two terms in this embodiment, one corresponding to the likelihood probability function and the other to the prior probability density function.
- the computation of the likelihood probability function gradient involves the application of standard image processing operations including geometric warping, a linear filtering, and sub-sampling/up sampling, for example, that model both the forward and the reverse of a low resolution image formation process.
- an initial, trial higher resolution image is needed. This may be, for example, a combination of one or more input (observed) lower resolution image sequences (block 30) that have been aligned (block 34) . The results of this initial alignment are then used with optical flow vectors 36 to repack the data (block 38) . Then the gradient is calculated (block 42) using an initial higher resolution image (block 40) and the repacked lower resolution data. Next, the initial higher resolution image is updated at block 44. The goal is to find the higher resolution image which, when re- sampled in the grid of the input lower resolution images, according to the imaging observation model, predicts the input (observed) lower resolution images.
- the gradients, computed in block 42 indicate to the iterative process the direction to move, so as to come closer to the peak or trough in the combination of likelihood and prior probability density functions.
- This movement along the likelihood and prior probability density functions results in a change or update (block 44) to the next higher resolution image, which generates the current trial higher resolution image.
- the current trial higher resolution image is then inserted into the second equation and compared to the threshold T (diamond 48) . If the delta E is still too high, then the gradient computation loop is repeated. This alignment may be evaluated using a conventional technique. Otherwise, if delta E is less than T, the sequence ends (block 50) .
- the lower resolution pixel data may be rearranged (i.e. repacked) before running the conjugate gradient optimization algorithm (Figure 3, block 38) .
- all of the low resolution pixels may be mapped to a certain area of a higher resolution image and all of the associated data may be put together in a continuous buffer.
- a continuous buffer is a circular buffer or a buffer that has no end.
- the image may be divided into tiles and then the gradient is computed per tile. Different tiles are processed in parallel with different threads in symmetric multiple processing (SMP) in a multicore processor.
- SMP symmetric multiple processing
- the higher resolution image is divided into a set of non-overlapping, rectangular tiles of the same or, substantially the same, size, as indicated in block 10.
- That higher resolution region may be a 3x3 region in one embodiment.
- the tile that the region center belongs to is found.
- the input lower resolution pixels, mapped into each tile, are then counted, as indicated in block 14 to get a count N t .
- a continuous buffer, pool[k] is allocated, as indicated in block 16.
- the size of the buffer is set equal to N t elements.
- the values of the relevant lower resolution pixels, together with the optical flow vectors, are placed in that continuous buffer, where the lower resolution pixel is mapped, as indicated in block 18.
- the optical flow vector is a two-dimensional vector with a physical offset across a two-dimensional image plane of this pixel between two frames.
- the gradients may then be calculated per tile, as indicated in Figure 2, using the reorganized data from the sequence of Figure 1.
- the gradient describes how far the current higher resolution image is from the estimated higher resolution image.
- the algorithm of Figure 2 may be executed in parallel for each tile in a multicore processor, in one embodiment. Each iteration runs separately and in parallel with the other iterations on a processor core assigned to a thread. In some embodiments, a processor with many cores may be utilized that is capable of running many threads.
- each loop indicated in Figure 2 processes all of the elements of the array assigned to the k-th tile.
- the elements of the array are the values of the optical flow vector .
- a delta is determined as the difference between a derivative of a robust function and a direct projection function (block 22) .
- the direct projection function projects the higher resolution image into lower resolution images. It models the mapping from pixels in the higher resolution image to points in the lower resolution images. The result is a set of lower resolution images .
- the gradient of that function is calculated.
- a term is determined for assessing the discrepancy between the observed lower resolution pixel value and the current (on the k-th minimization step) direct projection function value of the current higher resolution pixels, which is being estimated.
- a robust function may be used to find the optimal optical flow parameters of the direct projection function in some embodiments. Outliers are points that are affected by an unusually high amount of noise at the tail ends of the density of the probability density graph.
- One robust function is a Cauchy function which is equal to the logarithm of the quantity l+X 2 /sigma 2 , where sigma is a known parameter.
- the derivative of the robust function is 2*X/ (sigma 2 + X 2 ). Then, the gradient may be calculated by subtracting from the derivative of the robust function the value of the direct projection function.
- the next lower resolution pixel and optical flow vector are retrieved from a tile's continuous buffer.
- the direct projection of the higher resolution image into the lower resolution pixel is calculated using the retrieved optical flow vector.
- the gradient is updated with an inverse projected delta, as indicated in block 24.
- the inverse projection models mapping from pixels in the lower resolution image to pixels in the higher resolution image.
- the gradients calculated for each tile are combined into one gradient image.
- a graphics pipeline may include a graphics processor 112, coupled by a bus 106, to a frame buffer 114.
- a display screen 118 is coupled by a bus 107 to the frame buffer 114 and by a bus 108 to a keyboard or mouse 120.
- the graphics processor 112 may be coupled, in one embodiment, to chipset core logic 110 by a bus 105 and a main processor 100.
- one or more of the graphics processor or main processor may use symmetric multiprocessing with multiple cores to calculate gradients in parallel.
- a storage 130 may be coupled to the chipset core logic and the storage 130 may include a main memory 132, hard drives 134, and removable media 136, as examples.
- the sequences depicted in Figures 1, 2, and 3 may be implemented as machine readable instructions stored in a portion 139 of the main memory 132 in one embodiment.
- the continuous buffer, pool[k] may be established within the main memory 132.
- the buffer and the storage of the sequences, indicated in Figures 1 to 3 may also be implemented within the main processor 100, another memory, or in a graphics processor 112, to give additional examples.
- the sequences of Figures 1-3 may be implemented in software and, in other embodiments, they may be implemented in hardware.
- the sequences may be implemented by instructions executable by a processor, such as the processor 112 or the processor 100, in the cores of such a processor, using instructions stored in a computer readable medium, such as the region 139.
- the computer readable medium may be any computer readable storage medium, including magnetic, semiconductor, or optical memory.
- graphics processing techniques described herein may be implemented in various hardware architectures. For example, graphics functionality may be integrated within a chipset. Alternatively, a discrete graphics processor may be used. As still another embodiment, the graphics functions may be implemented by a general purpose processor, including a multicore processor.
- references throughout this specification to "one embodiment” or “an embodiment” mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation encompassed within the present invention. Thus, appearances of the phrase “one embodiment” or “in an embodiment” are not necessarily referring to the same embodiment.
- the particular features, structures, or characteristics may be instituted in other suitable forms other than the particular embodiment illustrated and all such forms may be encompassed within the claims of the present application.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
Super-resolution images may be produced by dividing a higher resolution image into a set of non-overlapping rectangular tiles of substantially the same size. Then, each pixel in each lower resolution image is mapped to the higher resolution image and it is determined which tiles are mapped to which lower resolution image pixels. A continuous buffer may be allocated for each tile and the relevant lower resolution pixels may be stored, together with optical flow vectors, in that continuous buffer. Then, the determination of gradients may use the information now stored in the buffer to facilitate symmetric multiprocessing using multi-core processors.
Description
COMPUTING HIGHER RESOLUTION IMAGES FROM MULTIPLE LOWER RESOLUTION IMAGES
Background
This relates generally to generating so-called super- resolution images which are higher resolution images based on multiple lower resolution images.
In electronic imaging applications, images with higher resolution are more desirable. Images with higher resolution have greater pixel density and, hence, show greater detail than lower resolution images of the same scene. Higher resolution images have many applications, including medical imaging, satellite imaging, computer vision, video surveillance, face recognition, car plate number extraction and recognition, and converting digital versatile disk video to high density television, to mention a few examples.
In super-resolution image reconstruction, multiple observed lower resolution images or frames of a scene are used to create a higher resolution image. The lower resolution images may be different views of the same scene. They may be obtained from the same camera, for example, while introducing small, so-called sub-pixel shifts in the camera location from frame to frame, or capturing a small amount of motion in the scene. Alternatively, the low resolution images may be captured using different cameras aimed at the same scene. A resultant high resolution image is then reconstructed by aligning and properly combining the low resolution images so that additional image information is obtained. The process may also include image restoration, where de-blurring and de-noising operations are performed as well.
The reconstruction of the resultant high resolution image is a difficult problem because it belongs to the class
of inverse, ill-posed mathematical problems. The needed signal processing may be interpreted as being the reverse of the so-called observation model, which is a mathematically deterministic way to describe the formation of low resolution images of a scene based upon known camera parameters. Since the scene is approximated by an acceptable quality high resolution image of it, the observation model is usually defined as relating a high resolution discrete image of the scene to its corresponding low resolution images. This relationship may be given as a concatenation of geometric transform, a blur operator, and down-sampling operator, plus an additive noise term.
Brief Description of the Drawings
Figure 1 is a flow diagram for a re-packing algorithm in accordance with one embodiment of the present invention; Figure 2 is a flow chart for the gradient computation in accordance with one embodiment of the present invention;
Figure 3 is a flow chart for non-linear iterative conjugate gradient optimization in accordance with one embodiment; and
Figure 4 is a system depiction for one embodiment.
Detailed Description
The observation model, a mathematically deterministic way to describe the formation of lower resolution images of a scene, may be expressed by the mathematical relationship Y=W*f+n, where Y is the set of observed lower resolution images and W represents the linear transformation of higher resolution pixels in the higher resolution image f to the lower resolution pixels in Y. The letter n represents additive noise having random characteristics, which may represent, for example, the variation or error between lower resolution images that have been captured by the same camera without any changes in the scene and without any changes to
camera or lighting settings. Based on the observation model, the super-resolution image reconstruction estimates the higher resolution image f that corresponds to a given set of lower resolution images Y. A Bayesian estimation process, sometimes referred to as stochastic or probabilistic super-resolution image reconstruction, may be used to estimate f, to get the resultant higher resolution image mentioned above. In that case, an "a posteriori" probability function, typically a probability density function, is mathematically defined as P(f|γ), which is the probability of a particular high resolution image f, given the set of observed low resolution images Y.
Applying a mathematical manipulation called Bayes law, the optimization problem, which is finding a suitable high resolution image f, that has the highest probability, given a set of low resolution images or that maximizes the a posteriori probability function, may be rewritten as: P(f |Y) = p(Y|f) * p(f), where p(f) is called the prior probability density function that gives the probabilities of a particular higher resolution image prior to any observation. The prior probability density function indicates what higher resolution images are more likely to occur, based on, for example, statistical characterization of an ensemble of different higher resolution images. The prior probability density function may be a joint probability, defined over all the pixels in a higher resolution image and may be based on statistical data from a large number of images. The a posteriori probability function, called the likelihood function, is a probability density function that defines the probabilities of observing lower resolution images that correspond to a particular higher resolution
image. The likelihood probability density function may be determined based on the observation model described above where the noise term is typically assumed to have a Gaussian probability distribution. The estimation process becomes one of iteratively determining trial higher resolution images and stopping when there is a convergence, which may signify that a maximum of the a posteriori probability function has been reached.
To define the maximum of the a posteriori probability function, a non-linear iterative conjugate gradient optimization algorithm is used. The conjugate gradient method computes the gradient of the a posteriori probability function at each search iteration. The calculation of the gradient is the most computationally intensive part of the super-resolution algorithm. In typical configurations it takes 80 to 90 percent of the maximization procedure running time .
The conjugate gradient optimization algorithm models noise by a probabilistic, non-Gaussian, robust function. Figure 3 shows the use of the conjugate gradient method which is an iterative method that provides an acceptable balance between complexity and speed of convergence. The criteria for convergence is delta E less than T (diamond 48), which tests whether the error or difference in the a posteriori probability function between two successive trial higher resolution images, is less than a predefined threshold T. An alternative is to define delta E as the difference between consecutive trial higher resolution images . The conjugate gradient method computes the gradient of the error function which has two terms in this embodiment, one corresponding to the likelihood probability function and the other to the prior probability density function. The
computation of the likelihood probability function gradient (block 38) involves the application of standard image processing operations including geometric warping, a linear filtering, and sub-sampling/up sampling, for example, that model both the forward and the reverse of a low resolution image formation process.
To compute the likelihood probability density function, an initial, trial higher resolution image is needed. This may be, for example, a combination of one or more input (observed) lower resolution image sequences (block 30) that have been aligned (block 34) . The results of this initial alignment are then used with optical flow vectors 36 to repack the data (block 38) . Then the gradient is calculated (block 42) using an initial higher resolution image (block 40) and the repacked lower resolution data. Next, the initial higher resolution image is updated at block 44. The goal is to find the higher resolution image which, when re- sampled in the grid of the input lower resolution images, according to the imaging observation model, predicts the input (observed) lower resolution images.
The gradients, computed in block 42, indicate to the iterative process the direction to move, so as to come closer to the peak or trough in the combination of likelihood and prior probability density functions. This movement along the likelihood and prior probability density functions results in a change or update (block 44) to the next higher resolution image, which generates the current trial higher resolution image. The current trial higher resolution image is then inserted into the second equation and compared to the threshold T (diamond 48) . If the delta E is still too high, then the gradient computation loop is repeated. This alignment may be evaluated using a
conventional technique. Otherwise, if delta E is less than T, the sequence ends (block 50) .
To reduce the memory footprint and to improve cache utilization, in some embodiments, the lower resolution pixel data may be rearranged (i.e. repacked) before running the conjugate gradient optimization algorithm (Figure 3, block 38) . In this rearrangement, all of the low resolution pixels may be mapped to a certain area of a higher resolution image and all of the associated data may be put together in a continuous buffer. As used herein, a continuous buffer is a circular buffer or a buffer that has no end. To achieve this rearrangement, the image may be divided into tiles and then the gradient is computed per tile. Different tiles are processed in parallel with different threads in symmetric multiple processing (SMP) in a multicore processor. Thus, referring to Figure 1, initially, the higher resolution image is divided into a set of non-overlapping, rectangular tiles of the same or, substantially the same, size, as indicated in block 10. For each pixel in each lower resolution image, its mapping to the higher resolution image is determined, as indicated in block 12. That higher resolution region may be a 3x3 region in one embodiment. Then, the tile that the region center belongs to is found. The input lower resolution pixels, mapped into each tile, are then counted, as indicated in block 14 to get a count Nt. For each tile k, where k is less than the total number of tiles and greater than or equal to zero, a continuous buffer, pool[k], is allocated, as indicated in block 16. The size of the buffer is set equal to Nt elements. The values of the relevant lower resolution pixels, together with the optical flow vectors, are placed
in that continuous buffer, where the lower resolution pixel is mapped, as indicated in block 18.
For each pixel in the lower resolution image, the optical flow vector is a two-dimensional vector with a physical offset across a two-dimensional image plane of this pixel between two frames.
After the initial data reorganization, the gradients may then be calculated per tile, as indicated in Figure 2, using the reorganized data from the sequence of Figure 1. The gradient describes how far the current higher resolution image is from the estimated higher resolution image. The algorithm of Figure 2 may be executed in parallel for each tile in a multicore processor, in one embodiment. Each iteration runs separately and in parallel with the other iterations on a processor core assigned to a thread. In some embodiments, a processor with many cores may be utilized that is capable of running many threads. Thus, each loop, indicated in Figure 2, processes all of the elements of the array assigned to the k-th tile. The elements of the array are the values of the optical flow vector .
After initialization (block 20), a delta is determined as the difference between a derivative of a robust function and a direct projection function (block 22) . The direct projection function projects the higher resolution image into lower resolution images. It models the mapping from pixels in the higher resolution image to points in the lower resolution images. The result is a set of lower resolution images . During the determination of a maximum of the a posteriori probability function on each iteration, the gradient of that function is calculated. To calculate the gradient, a term is determined for assessing the discrepancy
between the observed lower resolution pixel value and the current (on the k-th minimization step) direct projection function value of the current higher resolution pixels, which is being estimated. To minimize the influence of outliers, a robust function may be used to find the optimal optical flow parameters of the direct projection function in some embodiments. Outliers are points that are affected by an unusually high amount of noise at the tail ends of the density of the probability density graph.
There are a number of different robust functions. One robust function is a Cauchy function which is equal to the logarithm of the quantity l+X2/sigma2, where sigma is a known parameter. Thus, the derivative of the robust function is 2*X/ (sigma2 + X2). Then, the gradient may be calculated by subtracting from the derivative of the robust function the value of the direct projection function.
Thus, to calculate the delta, the next lower resolution pixel and optical flow vector are retrieved from a tile's continuous buffer. The direct projection of the higher resolution image into the lower resolution pixel is calculated using the retrieved optical flow vector.
Then, the gradient is updated with an inverse projected delta, as indicated in block 24. The inverse projection models mapping from pixels in the lower resolution image to pixels in the higher resolution image. Finally, as indicated in block 26, the gradients calculated for each tile are combined into one gradient image.
In some embodiments, the layout may be used in symmetric multiprocessing computers with multiple core processors. Referring to Figure 4, in accordance with some embodiments, a graphics pipeline may include a graphics processor 112, coupled by a bus 106, to a frame buffer 114.
A display screen 118 is coupled by a bus 107 to the frame buffer 114 and by a bus 108 to a keyboard or mouse 120.
The graphics processor 112 may be coupled, in one embodiment, to chipset core logic 110 by a bus 105 and a main processor 100. In some embodiments, one or more of the graphics processor or main processor may use symmetric multiprocessing with multiple cores to calculate gradients in parallel.
A storage 130 may be coupled to the chipset core logic and the storage 130 may include a main memory 132, hard drives 134, and removable media 136, as examples.
The sequences depicted in Figures 1, 2, and 3 may be implemented as machine readable instructions stored in a portion 139 of the main memory 132 in one embodiment. In addition, the continuous buffer, pool[k], may be established within the main memory 132. However, the buffer and the storage of the sequences, indicated in Figures 1 to 3, may also be implemented within the main processor 100, another memory, or in a graphics processor 112, to give additional examples.
In some embodiments, the sequences of Figures 1-3 may be implemented in software and, in other embodiments, they may be implemented in hardware. In a software embodiment, the sequences may be implemented by instructions executable by a processor, such as the processor 112 or the processor 100, in the cores of such a processor, using instructions stored in a computer readable medium, such as the region 139. The computer readable medium may be any computer readable storage medium, including magnetic, semiconductor, or optical memory.
The graphics processing techniques described herein may be implemented in various hardware architectures. For example, graphics functionality may be integrated within a
chipset. Alternatively, a discrete graphics processor may be used. As still another embodiment, the graphics functions may be implemented by a general purpose processor, including a multicore processor. References throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation encompassed within the present invention. Thus, appearances of the phrase "one embodiment" or "in an embodiment" are not necessarily referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be instituted in other suitable forms other than the particular embodiment illustrated and all such forms may be encompassed within the claims of the present application.
While the present invention has been described with respect to a limited number of embodiments, those skilled in the art will appreciate numerous modifications and variations therefrom. It is intended that the appended claims cover all such modifications and variations as fall within the true spirit and scope of this present invention.
Claims
1. A method comprising: forming a super-resolution image by mapping lower resolution pixels to a particular area of a higher resolution image; and collecting the associated data for said pixels within a buffer .
2. The method of claim 1 including collecting the associated data in a buffer that is continuous.
3. The method of claim 1 including dividing the higher resolution image into a set of non-overlapping rectangular tiles of substantially the same size.
4. The method of claim 3 including, for each pixel in each lower resolution image, finding its mapping to the higher resolution image.
5. The method of claim 4 including finding the tile within the higher resolution image that the pixels in the lower resolution image are mapped into.
6. The method of claim 5 including counting the number of lower resolution pixels mapped to each tile.
7. The method of claim 6 including, for each tile, allocating a continuous buffer of a size equal to the counted number of lower resolution pixels mapped into that tile.
8. The method of claim 7 including calculating a tile within the higher resolution image that a lower resolution pixel is mapped into and placing the value of the lower resolution pixels, together with its optical flow vector, within a continuous buffer for the tile.
9. The method of claim 8 including using the reorganized data to calculate a gradient for each tile.
10. An apparatus comprising: a multi-core processor to form a super-resolution image by mapping lower resolution pixels to a particular area of the higher resolution image; and a buffer coupled to said processor to collect the associated data for said pixels.
11. The apparatus of claim 10 wherein said buffer is continuous .
12. The apparatus of claim 10, said processor to divide the higher resolution image into a set of non-overlapping rectangular tiles of substantially the same size.
13. The apparatus of claim 12, said processor to find the mapping for each pixel in each lower resolution image to the higher resolution image.
14. The apparatus of claim 13, said processor to find the tiles within the higher resolution image that the pixels in the lower resolution image are mapped into.
15. The apparatus of claim 14, said processor to allocate a continuous buffer for each tile.
16. The apparatus of claim 15, said processor to place the values of the lower resolution pixels, together with optical flow vectors, within a continuous buffer for each tile.
17. The apparatus of claim 16, said processor to use the reorganized data to calculate a gradient for each tile.
18. The apparatus of claim 17, said processor to combine the gradients calculated for each tile into one combined gradient .
19. The apparatus of claim 18, said apparatus to calculate a difference between a derivative of a robust function and a direct projection function.
20. The apparatus of claim 19, said processor to update each gradient with an inverse projected delta.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/215,921 US8326069B2 (en) | 2008-06-30 | 2008-06-30 | Computing higher resolution images from multiple lower resolution images |
US12/215,921 | 2008-06-30 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2010002630A2 true WO2010002630A2 (en) | 2010-01-07 |
WO2010002630A3 WO2010002630A3 (en) | 2010-03-11 |
Family
ID=41165146
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2009/048241 WO2010002630A2 (en) | 2008-06-30 | 2009-06-23 | Computing higher resolution images from multiple lower resolution images |
Country Status (4)
Country | Link |
---|---|
US (2) | US8326069B2 (en) |
EP (1) | EP2141653A3 (en) |
CN (1) | CN101620730B (en) |
WO (1) | WO2010002630A2 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102011050585A1 (en) | 2011-05-24 | 2012-11-29 | Kunststoff Helmbrechts Ag | Method for producing plastic molded structure as display and/or functional element in motor vehicle, involves providing conductor track plane of lower functional film with electronic components, after molding of plastic molded structure |
CN105069765A (en) * | 2015-07-22 | 2015-11-18 | 广东迅通科技股份有限公司 | Fuzzy license plate reconstruction method based on characteristic learning |
US9950936B2 (en) | 2013-07-05 | 2018-04-24 | Mitsubishi Heavy Industries, Ltd. | Water treatment process and water treatment system |
Families Citing this family (82)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11792538B2 (en) | 2008-05-20 | 2023-10-17 | Adeia Imaging Llc | Capturing and processing of images including occlusions focused on an image sensor by a lens stack array |
EP2289235A4 (en) | 2008-05-20 | 2011-12-28 | Pelican Imaging Corp | Capturing and processing of images using monolithic camera array with hetergeneous imagers |
US8866920B2 (en) | 2008-05-20 | 2014-10-21 | Pelican Imaging Corporation | Capturing and processing of images using monolithic camera array with heterogeneous imagers |
WO2011063347A2 (en) | 2009-11-20 | 2011-05-26 | Pelican Imaging Corporation | Capturing and processing of images using monolithic camera array with heterogeneous imagers |
SG10201503516VA (en) | 2010-05-12 | 2015-06-29 | Pelican Imaging Corp | Architectures for imager arrays and array cameras |
CN102376086B (en) * | 2010-08-23 | 2014-03-26 | 致伸科技股份有限公司 | Method for generating high-resolution image |
US8878950B2 (en) | 2010-12-14 | 2014-11-04 | Pelican Imaging Corporation | Systems and methods for synthesizing high resolution images using super-resolution processes |
WO2012155119A1 (en) | 2011-05-11 | 2012-11-15 | Pelican Imaging Corporation | Systems and methods for transmitting and receiving array camera image data |
EP2726930A4 (en) | 2011-06-28 | 2015-03-04 | Pelican Imaging Corp | Optical arrangements for use with an array camera |
US20130265459A1 (en) | 2011-06-28 | 2013-10-10 | Pelican Imaging Corporation | Optical arrangements for use with an array camera |
WO2013043761A1 (en) | 2011-09-19 | 2013-03-28 | Pelican Imaging Corporation | Determining depth from multiple views of a scene that include aliasing using hypothesized fusion |
KR102002165B1 (en) | 2011-09-28 | 2019-07-25 | 포토내이션 리미티드 | Systems and methods for encoding and decoding light field image files |
EP2817955B1 (en) | 2012-02-21 | 2018-04-11 | FotoNation Cayman Limited | Systems and methods for the manipulation of captured light field image data |
US9210392B2 (en) | 2012-05-01 | 2015-12-08 | Pelican Imaging Coporation | Camera modules patterned with pi filter groups |
KR20150023907A (en) | 2012-06-28 | 2015-03-05 | 펠리칸 이매징 코포레이션 | Systems and methods for detecting defective camera arrays, optic arrays, and sensors |
US20140002674A1 (en) | 2012-06-30 | 2014-01-02 | Pelican Imaging Corporation | Systems and Methods for Manufacturing Camera Modules Using Active Alignment of Lens Stack Arrays and Sensors |
CN107346061B (en) | 2012-08-21 | 2020-04-24 | 快图有限公司 | System and method for parallax detection and correction in images captured using an array camera |
WO2014032020A2 (en) | 2012-08-23 | 2014-02-27 | Pelican Imaging Corporation | Feature based high resolution motion estimation from low resolution images captured using an array source |
US9214013B2 (en) | 2012-09-14 | 2015-12-15 | Pelican Imaging Corporation | Systems and methods for correcting user identified artifacts in light field images |
EP2901671A4 (en) | 2012-09-28 | 2016-08-24 | Pelican Imaging Corp | Generating images from light fields utilizing virtual viewpoints |
US9143711B2 (en) | 2012-11-13 | 2015-09-22 | Pelican Imaging Corporation | Systems and methods for array camera focal plane control |
WO2014130849A1 (en) | 2013-02-21 | 2014-08-28 | Pelican Imaging Corporation | Generating compressed light field representation data |
US9374512B2 (en) | 2013-02-24 | 2016-06-21 | Pelican Imaging Corporation | Thin form factor computational array cameras and modular array cameras |
US9638883B1 (en) | 2013-03-04 | 2017-05-02 | Fotonation Cayman Limited | Passive alignment of array camera modules constructed from lens stack arrays and sensors based upon alignment information obtained during manufacture of array camera modules using an active alignment process |
WO2014138697A1 (en) | 2013-03-08 | 2014-09-12 | Pelican Imaging Corporation | Systems and methods for high dynamic range imaging using array cameras |
US8866912B2 (en) | 2013-03-10 | 2014-10-21 | Pelican Imaging Corporation | System and methods for calibration of an array camera using a single captured image |
US9521416B1 (en) | 2013-03-11 | 2016-12-13 | Kip Peli P1 Lp | Systems and methods for image data compression |
US9106784B2 (en) | 2013-03-13 | 2015-08-11 | Pelican Imaging Corporation | Systems and methods for controlling aliasing in images captured by an array camera for use in super-resolution processing |
US9124831B2 (en) | 2013-03-13 | 2015-09-01 | Pelican Imaging Corporation | System and methods for calibration of an array camera |
WO2014164909A1 (en) | 2013-03-13 | 2014-10-09 | Pelican Imaging Corporation | Array camera architecture implementing quantum film sensors |
WO2014165244A1 (en) | 2013-03-13 | 2014-10-09 | Pelican Imaging Corporation | Systems and methods for synthesizing images from image data captured by an array camera using restricted depth of field depth maps in which depth estimation precision varies |
US9100586B2 (en) | 2013-03-14 | 2015-08-04 | Pelican Imaging Corporation | Systems and methods for photometric normalization in array cameras |
US9578259B2 (en) | 2013-03-14 | 2017-02-21 | Fotonation Cayman Limited | Systems and methods for reducing motion blur in images or video in ultra low light with array cameras |
US9445003B1 (en) | 2013-03-15 | 2016-09-13 | Pelican Imaging Corporation | Systems and methods for synthesizing high resolution images using image deconvolution based on motion and depth information |
US9497370B2 (en) | 2013-03-15 | 2016-11-15 | Pelican Imaging Corporation | Array camera architecture implementing quantum dot color filters |
US10122993B2 (en) | 2013-03-15 | 2018-11-06 | Fotonation Limited | Autofocus system for a conventional camera that uses depth information from an array camera |
US9497429B2 (en) | 2013-03-15 | 2016-11-15 | Pelican Imaging Corporation | Extended color processing on pelican array cameras |
US20140267759A1 (en) * | 2013-03-15 | 2014-09-18 | Digitalglobe, Inc. | Estimation of fine structure within pixels in satellite imagery |
US9633442B2 (en) | 2013-03-15 | 2017-04-25 | Fotonation Cayman Limited | Array cameras including an array camera module augmented with a separate camera |
WO2014145856A1 (en) | 2013-03-15 | 2014-09-18 | Pelican Imaging Corporation | Systems and methods for stereo imaging with camera arrays |
US9898856B2 (en) | 2013-09-27 | 2018-02-20 | Fotonation Cayman Limited | Systems and methods for depth-assisted perspective distortion correction |
WO2015070105A1 (en) | 2013-11-07 | 2015-05-14 | Pelican Imaging Corporation | Methods of manufacturing array camera modules incorporating independently aligned lens stacks |
WO2015074078A1 (en) | 2013-11-18 | 2015-05-21 | Pelican Imaging Corporation | Estimating depth from projected texture using camera arrays |
EP3075140B1 (en) | 2013-11-26 | 2018-06-13 | FotoNation Cayman Limited | Array camera configurations incorporating multiple constituent array cameras |
US9478010B2 (en) | 2013-12-12 | 2016-10-25 | Google Technology Holdings LLC | Generating an enhanced image of a predetermined scene from a plurality of images of the predetermined |
US10089740B2 (en) | 2014-03-07 | 2018-10-02 | Fotonation Limited | System and methods for depth regularization and semiautomatic interactive matting using RGB-D images |
US9247117B2 (en) | 2014-04-07 | 2016-01-26 | Pelican Imaging Corporation | Systems and methods for correcting for warpage of a sensor array in an array camera module by introducing warpage into a focal plane of a lens stack array |
CN103984937A (en) * | 2014-05-30 | 2014-08-13 | 无锡慧眼电子科技有限公司 | Pedestrian counting method based on optical flow method |
US10176452B2 (en) | 2014-06-13 | 2019-01-08 | Conduent Business Services Llc | Store shelf imaging system and method |
US10453046B2 (en) | 2014-06-13 | 2019-10-22 | Conduent Business Services, Llc | Store shelf imaging system |
US9659204B2 (en) | 2014-06-13 | 2017-05-23 | Conduent Business Services, Llc | Image processing methods and systems for barcode and/or product label recognition |
US9542746B2 (en) | 2014-06-13 | 2017-01-10 | Xerox Corporation | Method and system for spatial characterization of an imaging system |
US9521319B2 (en) | 2014-06-18 | 2016-12-13 | Pelican Imaging Corporation | Array cameras and array camera modules including spectral filters disposed outside of a constituent image sensor |
CN113256730B (en) | 2014-09-29 | 2023-09-05 | 快图有限公司 | System and method for dynamic calibration of an array camera |
US9942474B2 (en) | 2015-04-17 | 2018-04-10 | Fotonation Cayman Limited | Systems and methods for performing high speed video capture and depth estimation using array cameras |
KR102130361B1 (en) | 2015-12-03 | 2020-08-05 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Method and portable electronic device for changing graphics processing resolution according to scenario |
JP6596512B2 (en) * | 2015-12-25 | 2019-10-23 | 株式会社Pfu | Mobile terminal, image processing method, and program |
CN105677587A (en) * | 2016-01-04 | 2016-06-15 | 沈阳东软医疗系统有限公司 | Medical imaging equipment data storage method and device |
US9928438B2 (en) | 2016-03-10 | 2018-03-27 | Conduent Business Services, Llc | High accuracy localization system and method for retail store profiling via product image recognition and its corresponding dimension database |
CN109923579B (en) * | 2016-09-28 | 2024-01-26 | 港大科桥有限公司 | Restoration of pixel resolution in scanning imaging |
US10002344B2 (en) | 2016-10-17 | 2018-06-19 | Conduent Business Services, Llc | System and method for retail store promotional price tag detection |
US10210603B2 (en) | 2016-10-17 | 2019-02-19 | Conduent Business Services Llc | Store shelf imaging system and method |
US10019803B2 (en) | 2016-10-17 | 2018-07-10 | Conduent Business Services, Llc | Store shelf imaging system and method using a vertical LIDAR |
US10289990B2 (en) | 2016-10-17 | 2019-05-14 | Conduent Business Services, Llc | Store shelf imaging system and method |
RU2652722C1 (en) * | 2017-05-03 | 2018-04-28 | Самсунг Электроникс Ко., Лтд. | Data processing for super-resolution |
US10776904B2 (en) | 2017-05-03 | 2020-09-15 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image |
CN107292850B (en) * | 2017-07-03 | 2019-08-02 | 北京航空航天大学 | A kind of light stream parallel acceleration method based on Nearest Neighbor Search |
US10482618B2 (en) | 2017-08-21 | 2019-11-19 | Fotonation Limited | Systems and methods for hybrid depth regularization |
US10540749B2 (en) * | 2018-03-29 | 2020-01-21 | Mitsubishi Electric Research Laboratories, Inc. | System and method for learning-based image super-resolution |
CN108510446B (en) * | 2018-04-10 | 2023-03-14 | 四川和生视界医药技术开发有限公司 | Method and device for superimposing retinal images |
KR102646521B1 (en) | 2019-09-17 | 2024-03-21 | 인트린식 이노베이션 엘엘씨 | Surface modeling system and method using polarization cue |
EP4042101A4 (en) | 2019-10-07 | 2023-11-22 | Boston Polarimetrics, Inc. | Systems and methods for surface normals sensing with polarization |
EP4066001A4 (en) | 2019-11-30 | 2024-01-24 | Boston Polarimetrics, Inc. | Systems and methods for transparent object segmentation using polarization cues |
JP7462769B2 (en) | 2020-01-29 | 2024-04-05 | イントリンジック イノベーション エルエルシー | System and method for characterizing an object pose detection and measurement system - Patents.com |
WO2021154459A1 (en) | 2020-01-30 | 2021-08-05 | Boston Polarimetrics, Inc. | Systems and methods for synthesizing data for training statistical models on different imaging modalities including polarized images |
CN111367604A (en) * | 2020-02-28 | 2020-07-03 | Tcl移动通信科技(宁波)有限公司 | Processing method and device for file display, storage medium and terminal |
US11953700B2 (en) | 2020-05-27 | 2024-04-09 | Intrinsic Innovation Llc | Multi-aperture polarization optical systems using beam splitters |
CN113592709B (en) * | 2021-02-19 | 2023-07-25 | 腾讯科技(深圳)有限公司 | Image super processing method, device, equipment and storage medium |
US12020455B2 (en) | 2021-03-10 | 2024-06-25 | Intrinsic Innovation Llc | Systems and methods for high dynamic range image reconstruction |
US11290658B1 (en) | 2021-04-15 | 2022-03-29 | Boston Polarimetrics, Inc. | Systems and methods for camera exposure control |
US11954886B2 (en) | 2021-04-15 | 2024-04-09 | Intrinsic Innovation Llc | Systems and methods for six-degree of freedom pose estimation of deformable objects |
US11689813B2 (en) | 2021-07-01 | 2023-06-27 | Intrinsic Innovation Llc | Systems and methods for high dynamic range imaging using crossed polarizers |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080080261A1 (en) * | 2005-09-26 | 2008-04-03 | Rambus Inc. | Memory system topologies including a buffer device and an integrated circuit memory device |
US20080122946A1 (en) * | 2006-06-26 | 2008-05-29 | Samsung Electro-Mechanics Co., Ltd. | Apparatus and method of recovering high pixel image |
US20080143744A1 (en) * | 2006-12-13 | 2008-06-19 | Aseem Agarwala | Gradient-domain compositing |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6575367B1 (en) * | 1998-11-05 | 2003-06-10 | Welch Allyn Data Collection, Inc. | Image data binarization methods enabling optical reader to read fine print indicia |
US7016531B1 (en) * | 1999-02-01 | 2006-03-21 | Thomson Licensing | Process to extract regions of homogeneous color in a digital picture |
US7110027B2 (en) * | 2001-12-13 | 2006-09-19 | International Business Machines Corporation | Method and apparatus for producing still video images using electronic motion video apparatus |
US20080177994A1 (en) * | 2003-01-12 | 2008-07-24 | Yaron Mayer | System and method for improving the efficiency, comfort, and/or reliability in Operating Systems, such as for example Windows |
US7809155B2 (en) * | 2004-06-30 | 2010-10-05 | Intel Corporation | Computing a higher resolution image from multiple lower resolution images using model-base, robust Bayesian estimation |
US7447382B2 (en) | 2004-06-30 | 2008-11-04 | Intel Corporation | Computing a higher resolution image from multiple lower resolution images using model-based, robust Bayesian estimation |
JP3837575B2 (en) * | 2004-10-29 | 2006-10-25 | 国立大学法人東京工業大学 | Speeding up of super-resolution processing |
KR100781552B1 (en) | 2006-06-26 | 2007-12-05 | 삼성전기주식회사 | Apparatus and method for recovering high pixel image processing |
EP2052553A4 (en) * | 2006-07-26 | 2010-08-25 | Human Monitoring Ltd | Image stabilizer |
US8068700B2 (en) * | 2007-05-28 | 2011-11-29 | Sanyo Electric Co., Ltd. | Image processing apparatus, image processing method, and electronic appliance |
-
2008
- 2008-06-30 US US12/215,921 patent/US8326069B2/en not_active Expired - Fee Related
-
2009
- 2009-06-23 WO PCT/US2009/048241 patent/WO2010002630A2/en active Application Filing
- 2009-06-25 EP EP09251649A patent/EP2141653A3/en not_active Withdrawn
- 2009-06-30 CN CN2009101594364A patent/CN101620730B/en not_active Expired - Fee Related
-
2012
- 2012-11-06 US US13/669,597 patent/US8538202B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080080261A1 (en) * | 2005-09-26 | 2008-04-03 | Rambus Inc. | Memory system topologies including a buffer device and an integrated circuit memory device |
US20080122946A1 (en) * | 2006-06-26 | 2008-05-29 | Samsung Electro-Mechanics Co., Ltd. | Apparatus and method of recovering high pixel image |
US20080143744A1 (en) * | 2006-12-13 | 2008-06-19 | Aseem Agarwala | Gradient-domain compositing |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102011050585A1 (en) | 2011-05-24 | 2012-11-29 | Kunststoff Helmbrechts Ag | Method for producing plastic molded structure as display and/or functional element in motor vehicle, involves providing conductor track plane of lower functional film with electronic components, after molding of plastic molded structure |
US9950936B2 (en) | 2013-07-05 | 2018-04-24 | Mitsubishi Heavy Industries, Ltd. | Water treatment process and water treatment system |
CN105069765A (en) * | 2015-07-22 | 2015-11-18 | 广东迅通科技股份有限公司 | Fuzzy license plate reconstruction method based on characteristic learning |
CN105069765B (en) * | 2015-07-22 | 2017-12-22 | 广东迅通科技股份有限公司 | A kind of fuzzy license plate method for reconstructing of feature based study |
Also Published As
Publication number | Publication date |
---|---|
US8538202B2 (en) | 2013-09-17 |
US8326069B2 (en) | 2012-12-04 |
WO2010002630A3 (en) | 2010-03-11 |
EP2141653A2 (en) | 2010-01-06 |
US20090324118A1 (en) | 2009-12-31 |
CN101620730A (en) | 2010-01-06 |
EP2141653A3 (en) | 2012-09-12 |
US20130064474A1 (en) | 2013-03-14 |
CN101620730B (en) | 2013-05-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8326069B2 (en) | Computing higher resolution images from multiple lower resolution images | |
US9230303B2 (en) | Multi-frame super-resolution of image sequence with arbitrary motion patterns | |
WO2021208122A1 (en) | Blind video denoising method and device based on deep learning | |
Belekos et al. | Maximum a posteriori video super-resolution using a new multichannel image prior | |
US8768069B2 (en) | Image enhancement apparatus and method | |
KR101393048B1 (en) | Method and apparatus for super-resolution of images | |
Zhang et al. | A super-resolution reconstruction algorithm for surveillance images | |
JP5861949B2 (en) | Digital processing method and system for optical flow determination | |
US8958484B2 (en) | Enhanced image and video super-resolution processing | |
KR101137753B1 (en) | Methods for fast and memory efficient implementation of transforms | |
KR101252363B1 (en) | Method and apparatus for noise reduction in video | |
JP4741836B2 (en) | Robust estimation of camera pan vector by repeated use of center of mass | |
JP2008146643A (en) | Method and device for reducing blur caused by movement in image blurred by movement, and computer-readable medium executing computer program for reducing blur caused by movement in image blurred by movement | |
US20180005039A1 (en) | Method and apparatus for generating an initial superpixel label map for an image | |
Tahtali et al. | Graphics processing unit restoration of non-uniformly warped images using a typical frame as prototype | |
Messina et al. | Improving image resolution by adaptive back-projection correction techniques | |
Mohan | Adaptive super-resolution image reconstruction with lorentzian error norm | |
Al Ismaeil et al. | Multi-frame super-resolution by enhanced shift & add | |
Patel Shreyas | Novel Iterative Back Projection Approach | |
Pattanaik | Deep Learning based Super-Resolution for Medical Volume Visualization with Direct Volume Rendering | |
Lukeš | Super-Resolution Methods for Digital Image and Video Processing | |
Kyamakya et al. | Super Resolution Challenges and Rewards | |
CN117274029A (en) | Hardware acceleration method and system for LK optical flow algorithm | |
Gavade et al. | A Survey on Image Reconstruction Using Super Resolution | |
An et al. | Neighborhood-Based Weighted Regularization of Video Sequence Super-Resolution |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09774082 Country of ref document: EP Kind code of ref document: A2 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09774082 Country of ref document: EP Kind code of ref document: A2 |